International News on the Internet: Why more is less
Chris Paterson

This study examines the intersection of news agency political economy and cyberspace information exchange between 2001 and 2006 in an effort to determine if online news has corrected – or replicated – the inequities and limitations of the international journalism provided by ‘traditional media’. Through comparisons of content at major online news sites and the output of major news wholesalers, this research is the first to demonstrate that the international news most online users consult is that of just two news agencies. This research finds an online news environment with little real information diversity – a situation at odds with a decade and a half of fervour for the democratizing potential of new media.

Keywords: News agencies, online journalism, Reuters, associated press, news sources, international news

New media have often been deemed inherently democratizing and liberating, offering the prospect of freeing us all from a long standing dependence on a few powerful information providers and the “mainstream” discourse they offer. McNair (2003) hailed recent internet evolution and its infinite possibilities for horizontal communication, the demystification and deprofessionalization of journalism, and endless information choice, suggesting that the chaos of the contemporary communications environment may lay to rest concerns about the power of traditional media monoliths upholding the status quo. Bruns (2005) has argued that everyone can practice journalism and increasingly effective models of participatory online journalism are emerging, further eroding the power of “old media”. While each offer sophisticated improvements over utopian predictions common a decade earlier[1], there is reason to doubt if the long dominant providers of international affairs information have surrendered their privilege, or indeed, if they will.

This paper reports an exploratory study conducted in two phases between 2001 and 2006 to examine if online news has corrected – or replicated – the inequities and limitations of the international journalism provided by ‘traditional media’. This is an attempt to explore longstanding concerns about news flow and the possibility of cultural imperialism within the new media context, but it is an attempt limited by the difficulty of describing a rapidly changing phenomenon.

This project combines research into the recent history of the Internet news industry with longitudinal content analysis designed to measure the extent of source concentration and how it has changed during the period in question. This period has seen the emergence of the news aggregation industry and with it a disguised reliance on a surprisingly limited set of news organizations, even as that industry offers news consumers the illusion of information diversity and endless perspectives.

Dominance of a few major corporate websites

Would the pervasiveness of a few sources within the news pages of major websites preclude the possibility of audiences engaging with a wider range of news services and a greater diversity of information? The questions asked in most of the surveys on news consumption are too vague to definitively answer, but this study indicates that that most people using online news get most of their international news from just a few major corporate websites (as shown by industry ratings), and that those provide a very limited diet of information (as shown by the qualitative portion of this study). That is necessarily imprecise; as exploratory research this work should not be regarded as the final word on online news diversity, but should signal the need for further research which ties precise measurement of news consumption habits with measurement of news sources.

For simplicity, this research uses mostly US data, although the trends are international. By 2000, 23 per cent of Americans reported online news was their main news source, a figure which had increased ten percentage points in two years[2]. By 2004, about 42 per cent of Americans reported they ‘have at some time gotten news online’ (Pew 2005). Other surveys put the number as high as 60 per cent of US adults. According to the Pew Internet and American Life Project[3], ‘50 million Americans turn to the Internet for news on a typical day’. The Pew researchers argued: ‘News organizations have made news sites more attractive and rich with content in recent years.’ This research suggests that analyses of this sort are falling for a conjurer’s trick – being duped by more brand labels on the same very limited news content.

Surveys show a continuing loss of audience by television and print, but at a slower rate than at any time in the preceding decade (State of the News Media 2005). Determining precisely what online news audiences read is not straightforward. Ratings services determine time spent with the websites of leading companies, but generally do not report how much of that time was spent specifically with the news pages of those companies.

This project examines the period from 2001 to 2006, not because it is the ‘post-9/11’ era which many have pointed to as an era of significant change (Allan and Zelizer 2002), but because it is a period of maturation in the online news industry as represented by key milestones. These include the economic shakeout in the industry which occurred in 2001 as a result of the burst ‘dot com bubble’ – a short period which saw hundreds of online journalists made redundant and essentially the current shape of the industry emerge; the formulation of new media strategies by the leading news agencies, which have significantly fashioned the contemporary online news environment; and the development of models for online news distribution which have challenged and, in some cases, reinforced long standing trends in news flow.

Google News is especially significant in regard to the final point. This period has seen the widespread adoption of broadband, and studies have linked this with increased online news use (State of the News Media 2005; Pew 2005). It is also a period which has seen major news websites, such as MSNBC and the New York Times, move from being loss leaders, ‘willing to lose $200-300 million yearly’, and sustained by massive subsidy (McChesney 2000) to profit makers.

How ‘cybermediaries’ take an intermediary role

Sarker, Butler, and Steinfield (1995) questioned predictions regarding Internet commerce that intermediary information processors would fade from view given the opportunity in cyberspace for producers to reach consumers directly – the process of ‘disintermediation’. They reasonably suggested that ‘cybermediaries’ were taking the intermediary role. The early findings of the present research, based on analysis of internet news content in 1999 and 2001, indicated that this was mostly true in the international news arena, with a few news portals emerging as the dominant news cybermediaries, and original content producers accepting a less public role. Those studies showed that even the most prominent content producers for news online, like Time Warner (CNN Interactive) or Microsoft/General Electric (MSNBC), were, in regard to international news, mostly playing a cybermediary role – conveying with little editing or original journalism the news stories written by wire services. A crucial development of the past five years is that the online news sector is more evenly split between the cybermediaries of e-journalism (Paterson 1999) and disintermediated producers of original content distributing content directly to consumers – in accordance with those earliest predictions about internet commerce.

That is one layer of a more complex cake. Another is the phenomenon of a mature news aggregation industry, containing sites such as Yahoo, Altavista, Google and Excite, where a few original producers of content provide the lion’s share of the international news for those aggregators, despite the audacious pretence of source diversity which each promotes. According to various ‘ratings’ reports of online use, most of the online news audience spends most of their time with a small number of websites, mostly in the guise of news aggregators, and this study confirms that those sites mostly relay news from the same few sources.

By 2001, it was apparent that the dominant online news providers were taking two main forms. The first were on-line media consisting mostly of the portal sites of major conglomerates, relaying mostly news agency content to audiences. A portal is a website designed to serve as a web user’s home page and primary contact point with the internet. It is intended to make audience eyes ‘stick’ to the advertising sold by the portal’s owner. Yahoo was the first to develop a strategic relationship with the Reuters news agency in the mid-1990s to facilitate such sites, and their model was widely copied. The second type were content-producing on-line media, which tended to combine original content (written by the company’s own journalists or commissioned for the company) with news agency content. That category included traditional media like the BBC or the New York Times, along with news companies which began on the web (Nando Times or Out There News are well known examples).

By the end of the period in question, the term ‘news aggregator’ had come into wide use, although it is deceptive. The aggregators may be portals or search engines which have developed mechanisms for retrieving, selecting, ranking, and linking to a massive amount of news posted to other parts of cyberspace. Some, like Google News, began this way, while others, like Yahoo, evolved from being mostly an online outlet for the wire services to an aggregator of news from what appear to be thousands of news outlets. The question of appearance is crucial, though, for this research demonstrates that despite such apparent choice international news still comes from few sources – the international news agencies.

Online news distribution

The portals, aggregators, and search engines which are the leading online news providers have substantially merged into the same thing. In the last five years portals have transformed into aggregators to make the apparently enormous amount of news on the internet easily available to their audience – while holding that audience. There are two key distinctions. One is between sites that produce original content and those that don’t, although Yahoo especially is blurring that distinction; and the other is between those with human control over news selection and those pioneering automated selection, of which Google News is the leader. Google News was created in 2002 amid great hype about its ‘automated editing’ but the service provides a range of news on multiple topics just as the more established online services did; it just provides far more of it.

When compared to the news selection underway at other mainstream news websites, Google News has been found to consistently pick the same stories (White 2003)[4]. Is this evidence of the genius of Google’s algorithms, or evidence that journalists everywhere determine news importance more or less identically? Could it be evidence that the human editors at Yahoo and CNN are looking, from time to time, at other popular news services to see how they rank stories, and adjusting their rankings accordingly? Or, as this article posits, could it be that the news agenda of all the popular online news services is substantially determined by the similar choices of two wire services? All are possibilities, but it is beyond the scope of this research to provide definitive answers. Google News has yet to rival the more established portals. According to Nielsen/Netratings data, in 2004 Google News reached 6.3 million people, far below the news audience of the other major online news services (Hearn 2005), and so it is not included in the quantitative portion of this study.

Google’s approach riled the news agencies. In 2005, Agence France-Presse (AFP) announced a lawsuit in the United States and France against Google for continuously breaching their copyright. The suit was settled in 2007 when the companies signed a licencing agreement. Several months earlier Google had agreed to pay the Associated Press for its content, although the cost to Google of both agreements, and how they differ from news agency agreements with other online news providers, is not known. Other aggregators, including Yahoo and AOL, pay AFP for the use of content (Jones 2006). Yahoo claims to pay some news services for content, including the major agencies, but to use many others “informally” (Washington Internet Daily 2006). One industry newsletter reports that most aggregators do not pay for content (ibid). The World Association of Newspapers announced that it would investigate ‘options regarding copyright infringement by search engines and news aggregates’ (ibid). Google News has been found guilty of copyright infringement in Belgium in a suit brought by newspapers (Blenkinsop 2007), raising the possibility of massive fines and leaving uncertain the long-term viability of the Google News model.

Increasing concentration

The analysis of what news a few online companies offer audiences[5] would be less consequential if hopes for the internet had been fulfilled. If news readers pursued a wide range of sources, as they can, the focus on a few would be, at least, less urgent. But the online news audience has demonstrated that it will not behave according to utopian predictions. Instead, it behaves as it always had with old media – it identifies (with the guidance of powerful marketing directed its way) a few favourite channels of information, and develops a loyalty to these that is extraordinary in view of the potential for taking in a wider view of the world.

Industry surveys of online use have demonstrated that Internet users spent their time with the websites of ever fewer corporations. By 2001 MediaMetrix was reporting that US web users spend more than 50 per cent of their time online with websites owned by four companies: AOL, Yahoo, Microsoft, and Napster (CNN 2001; Solomon 2001). An analyst from MediaMetrix, suggested these results ‘show an irrefutable trend toward online consolidation and indicate that the playing field is anything but even’ (CNN 2001). The top news sites correspond almost precisely to the top media companies worldwide.

Despite the ability of international conglomerates to dominate online news since the earliest days of the World Wide Web, there seemed a time that the future of their news operations was in doubt. As the ‘dot com bubble’ burst in the 2001, major firms eliminated efforts at original journalism, giving rise to the dominant aggregation model. In 2005 ‘62 per cent of Internet journalists said their newsrooms have suffered recent cutbacks, almost twice the 37 per cent of national print, TV and radio journalists to report that their newsrooms have suffered cutbacks’ (State of the News Media 2005). The trend since 2001 has been away from investment in online news; paradoxically, this as occurred as the quantity and scope of online news providers has increased.

In 2001, it was reported that one third of the time US net users spent online was with AOL-Time Warner websites. When confined to home, versus business, use, the number rose to 75 per cent. A 2003 Nielson/Netratings ranking of the top twenty ‘Current Events and Global News Sites’ (in the US) showed over a quarter (26 per cent) of the audience using Time Warner websites (Netscape, AOL, CNN, Time), and roughly a sixth going to both Yahoo (Yahoo! News) and Microsoft sites (MSNBC, Slate). Other leading sites were owned by the Washington Post (which would take over Slate in 2004), Disney, the New York Times, and News Corporation. A 2006 Pew foundation survey found that 46 per cent of US Internet users go to the website of a national TV news company such as CNN or MSNBC, while 39 per cent go to ‘portal websites such as Yahoo or Google’ (Horrigan 2006). In each case the percentages are slightly higher for Internet users with broadband connections (ibid).

The following were reported in mid-1996 to be the leading websites providing international news in the US (Nielsen 1996), with the number of monthly unique visitors:
Corporation Audience
Microsoft 114,293,000
Yahoo! 106,619,000
Time Warner 102,681,000
Google 95,340,000
News Corp. Online 61,752,000

Only one of these services is essentially just news – the New York Times (although it too has become heavily entertainment oriented). Companies like Microsoft will know from their internal research how much time their monthly audience is spending with their journalism services, but there seems no public data to reveal this. In the UK, the top websites are similar (compiled from earlier data and multiple ratings services):
Corporation Audience
Microsoft 20,782,000
Google 19,095,000
Yahoo! 13,384,000
BBC 11,541,000
Time Warner 8,331,000
News Corp. Online 6,422,000

Hitwise (2006) reported that in the UK, the BBC is the most popular news site, capturing just under 40 per cent of online news users (shared, apparently, between news.bbc and bbc.co.uk). It is followed by the Guardian Unlimited, Google UK News, CNN.com, Yahoo News, Times Online, and the Telegraph. The leading news providers are almost identical as well in Germany, with the addition of Bertelsmann as a major news provider, with sites like Stern and RTL Television.

The changing news agency role

While research on news agencies is limited, researchers have described how increasing concentration of control over the global wholesale news system made the global news agencies more influential than they had ever been, and the leading two are the New York- based Associated Press (AP) and London-based Reuters (recently acquired by the Thomson conglomerate and re-dubbed Thomson-Reuters)[6]. That is mostly the result of television networks of the wealthiest nations curtailing their reporting since the 1980s, and relying more on agencies as a result; that, in turn, was mostly the result of the determination of new corporate owners like Disney and General Electric that news divisions should pay their own way[7].

Because news agencies must please all editors, everywhere, they work harder than their client news organisations to appear objective and unbiased. The result is a bland and homogeneous, but ideologically distinctive, view of the world; stories challenging the ideological positions of dominant global political players (in agency eyes, the US and UK) receive little attention (Paterson 2005). News agency research has demonstrated content dictated by the ideological, structural, and cultural nature of these organisations (Paterson 1996; Cohen et al 1996; Hjarvard 1995; Wallis and Baran 1990). But importantly, agencies set the agenda for what international stories other media carry, through the choice of stories they distribute and the amount of visuals provided (moving for TV, still for newspapers and magazines, and both for webcasters), and in the case of agency-provided TV pictures, the nature and amount of accompanying audio and textual information[8]. Global and regional news agencies have grown more crucial as they bypass intermediary processors of news in cyberspace enabling them to directly reach – for the first time – a large portion of the news audience.

At the beginning of the period in question, dependence on AP or Reuters stories provided news websites an important association with well known and trusted, if little understood, news brands. Each was a vaguely known news brand without the negative associations familiar media outlets might have had, for the mass audience had little direct experience with agencies which could create such associations. A similar dependence on the two largest agencies was also driven by the need to have the same stories rival sites had. Thus, near identical menus of world stories would appear on each major site daily, based on agency news selection.

While Reuters and the Associated Press are equally ubiquitous in cyberspace, they have pursued different online strategies. Reuters aggressively moved away from its roots it terms of distribution, while the AP has mostly remained tied to the subscription model it has relied on for 150 years. Consistent with its origins, the AP, as a means of protecting the print media which own it, only provides links at its website to its content through the websites of member newspapers. Reuters still depends upon the subscription model in agreements with the media organisations to which it feeds video, audio, and text; but it has – in the last five years – also gone into competition with its subscribers and become an online news service, marketing directly to the consumer and provided its branded stories to news aggregators.

Measurement of agency use

The quantitative portion of this analysis employs a simple content comparison to determine the amount of verbatim news agency use by major news websites within new international stories. The rationale is to provide a previously non-existent measure of which news organisations are providing original news reporting of global events which fall outside of a select few ‘mega-stories’ (such as the Iraq war). A small amount of average verbatim news agency use is an indication that a news website is drawing from its own international resources and consulting a range of sources to construct original and unique stories, whereas a high degree indicates little investment or effort toward original journalism, leading to a subsequent lack of diverse perspectives on international events.

At the outset of this project, a pilot study was conducted with a single story in 1999. For that analysis, a single, relatively fast breaking and important (in the sense of gaining worldwide attention) story was selected. The story was the escalation of violent student rioting in Tehran on the morning of July 12, 1999. Over a period of just a few hours, the text of stories published on-line by each of the following services were copied and electronically stored: Reuters, Associated Press, Agence France-Presse, Yahoo, MSNBC, ABC Online, CNN Interactive, and BBC Online[9].

It proved difficult to precisely quantify news agency use, but it was easy to demonstrate that major online news services produced almost no original journalism in this case, and published stories that were almost entirely barely-edited wire service material. For example, analysis of the CNN, BBC and MSNBC stories revealed that less than five paragraphs from these three services combined (comprising 38 paragraphs in total) were not close or exact duplications of paragraphs written by wire services (that is, original compositions possibly involving reporting from other sources). It is important to note that within the news industry such a finding is neither remarkable nor alarming. It has always been the role of the agencies to provide the words when a news service does not have their own correspondent on the scene, and the contracts the agencies provide to clients entitle them to use agency copy in this way. But this dependence was surprising in view of the pretensions of these news outlets to be international news services in their own right. This data suggested a lack of investment in original online international reporting which called for investigation.

The study was repeated with a sample of fourteen international stories in 2001. The researcher compared the texts data set using a basic text comparison software called ‘Copyfind’, developed to detect plagiarism[10]. Comparisons were made between each wire service story and the online news story published (or linked to) by the news services in the sample[11]. A total word count was performed for each sample of news text. In a typical result, for a 642 word CNN story on UN troops in the Congo, 553 words existed in phrases (strings of five words or more) copied from Reuters, and 29 words existed in phrases copied from AP. This was, in other words, a virtually unchanged Reuters story published by CNN (although CNN did not identify it as such). Finally, the amount of text each news service copied from news agencies were compared to the total amount of text produced by that service to provide the percentage of news agency use given below. The study was repeated in 2006, with a slightly differing line-up of news organisations. In each case, the researcher included major US and UK online media, and both print and broadcast websites, within the limits of the most popular news brands identified earlier[12].

There are many complications in any attempt to track sources of online news. Not least of these is the inconvenient fact that both the news agency copy and each news service story based upon it are moving targets – prone to constant change. Minor updates of existing online stories are know in the industry as ‘write-throughs’, and are a logical extension of long standing news agency practice. News agencies have, since the earliest days of the telegraph, sent first the bare facts, and then progressively added details and corrected information that was originally wrong. The result is a series of updates after a story is first released, which may continue for hours or days after the first version.

When updating is practiced by news organizations whose primary client is the news consumer, not other journalists, the practice becomes ethically thorny. In newspapers and broadcast, journalists do not get the opportunity to continually improve and modify the work they publish, and readers and viewers reasonably expect the story they see to be complete and accurate when they see it. But online journalists have proven comfortable with making frequent alterations to their work as new information comes in or old information is brought into question, or at times, when editors decide upon a more appropriate framing.

In an innovative computer-assisted content analysis, Kutz and Herring (2005) determined that ‘that the second most common type of revision (after clarification) adds ideology’. As with this author’s 1999 research, these researchers found that the news services they observed routinely rewrote stories to add ‘more emotionally manipulative’ words and phrases.[13]

Results

The basic results of the 2001 and 2006 content surveys are provided below[14].

Average Percentage of Verbatim News Agency Use, by Online News Service

| |AOL |YAHOO |NANDO |LYCOS |EXCITE |ALTAVISTA | |Row Total |
| | | | | | | | |Avg. % |
|2001 |88 |69 |41 |87 |89 |36 | |68 |
|2006 |94 |97 | | |98 |50 | |85 |
| | | | | | | | | |
| |MSNBC |CNN |BBC |ABC |SKY |GUARDIAN |NYT | |
|2001 |53 |36 |5 |55 |6 | |47 |34 |
|2006 |81 |59 |9 |91 |15 |62 |32 |50 |

In 2001, news portals/aggregators showed substantially no mediation of agency content, with their text duplicating news agency text for an average of 68 per cent of the content studied (the average of the average duplication for each service)[15]. By 2006, the average amount of measurable verbatim news agency use for these services had risen to 85 per cent. While AFP was included, almost all the news agency use is split between Reuters and Associated Press. The limitations of the methodology and small sample mean these figures are at best indicative, but they strongly suggest an increasing and now near total dependence on two news agencies among the top row of online sources, the aggregators.

Although a central purpose of this research was to find news agency content where it is not acknowledged, in some instances the researcher had to depend on acknowledgement of agency sources by news services themselves[16]. In other cases, the main or only story at some portals was that of the New York Times or another non-agency source. In such cases the portal’s editors seem to have taken the decision that the Time’s coverage was original and superior to that of the news agencies. Such cases were rare within this sample.

The bottom row of data relates news agency dependence indicators for the major original news content providers – the popular ‘old media’ web outlets. The figure has risen from 34 per cent dependence to 50 per cent in five years. It is noteworthy that the major US sites, CNN, MSNBC, and ABC, seem to be doing substantially less original international journalism than they once were. The New York Times seems to be doing more. BBC Online continues to exhibit little verbatim use of news agencies, and stories generally appear to draw from a wider range if sources than the American services. Judging from the few Sky stories in the sample, Sky appears to draw just from news agencies, but their text has been extensively reworded in Sky’s concise tabloid-like style.

It is thus possible to reach the general conclusion that international news flow on the Internet has increased in apparent diversity of original reporting but decreased or remained static in actual diversity of original reporting. This longitudinal comparative analysis of international stories from major news web sites with original wire service stories reveals a continuing scarcity of original journalism (or even copy-editing) and a nearly total dependence by major online news providers on news agency reporting and writing. There seems a shift in these five years toward less minor rewriting of wire stories, with the broadcaster-based services opting more often now to simply publish wire stories in their entirety and clearly put the wire services label on the stories[17]. In effect, they seem less concerned with the image of providing original news coverage and more concerned with providing a massive quantity of coverage.

Conclusions

We are left with a picture of an online news world (in the English language) where only four organizations do extensive international reporting (Reuters, AP, AFP, BBC) a few others do some international reporting (CNN, MSN, New York Times, Guardian and a few other large newspaper and broadcasters), and most do no original international reporting. It makes the news aggregation industry appear somewhat inane – why not just link to the four companies filing original reports from around the world and ignore the rest? That is not an acceptable solution, of course, either in terms of the marketing priorities of the aggregators, nor for the global public sphere.

This research indicates that discourse on international events of consequence within the global public sphere is substantially determined by the production practices and institutional priorities of two information services – Reuters and the Associated Press[18]. The political economy of online news is not one of diversity but one of concentration, and the democratic potential of the medium remains that – potential.

It makes economic sense that the two leading news agencies should dominate international news delivery in cyberspace, for as in any open and unregulated market, the strongest producers with the lowest unit costs thrive. Such is the case for the major wire services, which each have a century and a half of experience in developing production processes which generate massive amounts of news. Digital technologies have made news agency production more efficient, and their (technological) convergence has permitted easy access into new markets through the creation of products tailored to new media, built from the same agency words and pictures upon which traditional media have long depended.

In the context of international news, the internet has transitioned into what we have traditionally regarded as ‘old media’: it is now, for most users, a mass medium providing mostly illusory interactivity and mostly illusory diversity. It is becoming a substantially tailored mass media product through the personalisation of information delivery, but these phenomena make it no less a form of mass media than would the insertion of targeting advertising into a magazine delivered to someone’s home. Because resources are being devoted to endless distribution and redistribution, internet journalism will continue to grow thinner. Given the massive explosion of distribution, there is surprisingly little new original journalism within the mainstream (mass audience) worldwide web. Whether or not the blogging phenomena offers a solution might be a point of contention, but research on blogging to date indicates that for remote international stories, bloggers are as dependent on the sources described here as everyone else.

The evolution of online news has laid bare the online industries’ dependence on a few wholesale providers. Such concerns were destined to remain substantially academic until the news industry itself revealed its previously hidden wholesale-retail structure online for the mass audience. While the online news industry continues to pretend for the moment that it brings readers a diversity of reporting on world news, it is a pretence which cannot last. And the ethical implications of maintaining that pretence are worthy of greater analysis.

It is likely that in the near term the online news industry will try harder to disguise its dependence on limited sources through cosmetic change, the addition of minor editorial adjustments to agency stories (by machine and human), and the addition of further bells and whistles at news sites. They seek to distract users from the problem. In the longer term the industry must invest in original reporting as an alternative to the few genuinely international news organizations now on offer, and give more prominence to buying, and properly translating, original non-English language reporting from around the world. Without such change, new media will continue to present to most users the dangerous illusion of multiple perspectives which actually emanate from few sources.

References

Allan, S. and Zelizer, B. (2002) Journalism after September 11, London, Routledge

Aufderheide, P. (1998) Niche-market culture, off- and online, in Borden, D. and Harvey, K. (eds) The electronic grapevine: Rumor, reputation, and reporting in the new on-line environment, Mahwah, NJ, Lawrence Erlbaum

BBC (1999) Web is ‘shrinking’, August 25. Available online at http://news.bbc.co.uk/hi/english/sci/tech/newsid%5F428000/428999.stm, accessed April 2006

BBC (2001) CNN to axe 400 jobs, January 18. Available online at http://news.bbc.co.uk/1/hi/world/americas/1121781.stm, accessed April 2006

BBC (2001) Disney's internet division closed, January 30. Available online at http://news.bbc.co.uk/1/hi/business/1143425.stm, accessed April 2006

BBC (2001) Yahoo links with New York Times, February 13. Available online at http://news.bbc.co.uk/1/hi/business/1168993.stm, accessed April 2006

Benz, B. and Phillips, M. (2005) Time for a change: The Associated Press as Napsterized news, Online Journalism Review, April 28. Available online at http://www.ojr.org/ojr/stories/050428benzphillips/, accessed April 2006

Blenkinsop, P. (2007) Belgian court rules against Google over copyright, Reuters.com, February 13 accessed April 2007

Blevins, J. L. (2004) Battle of the online brands: Disney loses Internet portal war, Television and New Media, Vol. 5, No. 3

Borden, D. and Harvey, K. (1998) The electronic grapevine: Rumor, reputation, and reporting in the new online environment, Mahwah, NJ, Lawrence Erlbaum

Boyd-Barrett, O. (2000) National and international news agencies: Issues of crisis and realignment, Gazette, Vol. 62. No. 1 pp 5–18

Boyd-Barrett, O. (1980) The international news agencies, London, Constable

Boyd-Barrett, O. and Rantanen, T. (2004) News agencies as news sources: A re-evaluation, in Paterson, C. and Sreberny, A. (eds) International news in the twenty-first century, Eastleigh, John Libbey

Boyd-Barrett, O. and Rantanen, T. (1998) The globalization of news, London, Sage

Bruns, A. (2005) Gatewatching: Collaborative online news production, New York, P. Lang

Burden, P. (1999) Interactivity and on-line news at the BBC, unpublished Masters dissertation, CMCR, Leicester, University of Leicester

Burns, E. (2006) Top U.S. parent companies and stickiest brands on the web, February 2006, clickz.com, March 27. Available online at http://www.clickz.com/stats/sectors/traffic_patterns/article.php/3594576#table1, accessed April 2006

Christopher, L. (1998) Technology and journalism in the on-line newsroom, in Borden, D. and Harvey, K. (eds) The electronic grapevine: Rumor, reputation, and reporting in the new on-line environment, Mahwah, NJ, Lawrence Erlbaum

CNN (2001) Study: Four sites account for half of Web surfing, June 5. Available online at http://archives.cnn.com/2001/TECH/internet/06/05/internet.consolidation/index.html, accessed April 2006

Cohen, A., Levy, M., Roeh, I. and Gurevitch, M. (1996) Global newsrooms, local audiences: A study of the Eurovision News Exchange, Luton, John Libbey

Cozens, C. (2005) AFP sues Google over copyrighted content, Guardian, March 21. Available online at http://media.guardian.co.uk/newmedia/story/0,7496,1442703,00.html, accessed April 2006

Delaney, K. (2005) Yahoo ‘Hybrid’ Now Dominates News Web Sites, Wall Street Journal, April 14. Available on ABI/INFORM

Fenby, J. (1986) The international news services, Twentieth Century Fund Report, New York, Schocken Books

Harper, C. (1997) The daily me, American Journalism Review, April.

Hearn, K. (2005) Google’s bias for bigness, Alternet, July 14. Available online at http://www.alternet.org/story/23397, accessed April 2006

Helland, K. (1995) Public service and commercial news, doctoral thesis, University of Bergen

Hills, J. and Michalis, M. (2000) The Internet: A challenge to public service broadcasting? Gazette, Vol. 62, No. 6

Hitwise (2006) BBC favoured news source but Wikipedia and Flickr growing in importance in the UK, January 5. Available online at http://weblogs.hitwise.com/heather-hopkins/2006/01/bbc_favoured_news_source_but_w.html, accessed April 2006

Hjarvard, S. (1995) Internationale TV-nyheder. En historisk analyze af det europeiske system for udvkesling af internationale TV-nyheder, Copenhagen, Akademisk Forlag

Hjarvard, S. (1995) TV news flow studies revisited, Electronic Journal of Communication, Vol. 5 No. 2

Horrigan, J. (2006) For many home broadband users, the internet is a primary news source, Pew Internet and American Life Project, March 22, Available online at http://www.pewinternet.org/pdfs/PIP_News.and.Broadband.pdf, accessed April 2006

Hu, J. and Olsen, S. (2002) Cutbacks push dot-coms toward profits, CNET, January 9. Available online at http://news.com.com/Cutbacks+push+dot-coms+toward+profits/2100-1023_3-804066.html, accessed April 2006

Jones, M. (2006) AFP sues Google over copyrights, Presstime. Available on LexisNexis

Kempner, M. (2001) CNN gains: Web site closes in on MSNBC.com, Atlanta Journal-Constitution, June 22. Available on LexisNexis

Kutz, D. and Herring, S. (2005) Micro-longitudinal analysis of web news updates, in Proceedings of the 38th Hawaii International Conference on System Sciences

Langfield, A. (2002) Net news lethargy most sites fail to make use of the medium’s main strength – speed, Online Journalism Review. Available online at http://www.ojr.org/ojr/reviews/1017864558.php, accessed May 2003

Liedtke, M. (2005) AP to impose online licensing fees, Associated Press Financial Wire, April 19. Available on LexisNexis

Lipscomb, G. (2001) BBC Online put under scrutiny, Broadcast, January

Madore, J. (2005) The Associated Press drops plans for fee, Newsday, July 23. Available on LexisNexis

Magid, L. (1997) All roads lead to Reuters, Currents.net. Available online at http://www.currents.net/magazine/national/1513/uout1513.html, accessed May 2003

Martinson, J. (2004) Reuters takes its news into the home, Guardian, October 6.

Massey, B. and Levy, M. (1999) Interactivity, on-line journalism, and English-language web newspapers in Asia, Journalism and Mass Communication Quarterly, Vol. 76, No. 1

McChesney, R. (2000) The Titanic sails on: Why the Internet won’t sink the media giants, Extra! March/April

McNair, B. 2003, ‘From control to chaos: towards a new sociology of Journalism’ Media, Culture and Society, Vol. 25 pp 547–555

Molina, G. (1990) The production of Mexican television news: The supremacy of corporate rationale, unpublished doctoral thesis, University of Leicester

Nielsen-NetRatings (2006) User-generated content drives half of U.S. top 10 fastest growing web brands, according to Nielsen//Netratings, Netratings press release, August 10. Available online at http://www.netratings.com/press.jsp?section=ne_press_releases&nav=1#, accessed May 2007

Paterson, C. (1996) News production at Worldwide Television News (WTN): An analysis of television news agency coverage of developing countries, doctoral thesis, University of Texas

Paterson, C. (1998) Global battlefields, pp 79-103

Paterson, C. (1998) Global news agencies, in Boyd-Barrett, O. and Rantanen, T. (eds) The globalization of news, London, Sage

Paterson, C. (1999) Internet news: Source concentration and cybermediation, EURICOM Colloquium on the Political Economy of Convergence, London

Paterson, C. (2001) Media imperialism revisited: The global public sphere and the news agency agenda, in S. Hjarvard (ed.) News in a globalized society, Göteborg, NORDICOM

Paterson, C. (2004) Shouting from the rooftops: Reflections on International News in the 21st Century, in Paterson, C. and Sreberny, A. (eds) International News the Twenty-First Century, Eastleigh, John Libbey Press

Rodriguez, A. (1996) Made in the USA: The production of the Noticiero Univision, Critical Studies in Mass Communication, Vol. 13, No. 1

Ross, S. and Middleberg, D. (1998) Media in cyberspace study. Available online at http://www.middleberg.com/toolsforsuccess/cyberstudy.cfm, accessed May 2003

Rubenstein, D. (2002) Google News untouched by human hands, Vue Weekly via AlterNet, October 7. Available online at http://www.alternet.org/story/14239, accessed April 2006

Sarker, M., Butler, B., and Steinfield, C. (1995) Intermediaries and cybermediaries: A continuing role for mediating players in the electronic marketplace, Journal of Computer-Mediated Communication No. 54. Available online at http://www.ascusc.org/jcmc/vol1/issue3/sarker.html, accessed February 2001

Scott, B. (2005) A contemporary history of digital journalism, Television and New Media, Vol. 6, No. 1

Solomon, N. (2001) Denial and the ravaging of cyberspace, AlterNet, September 4. Available online at http://www.alternet.org/columnists/story/11422, accessed April 2006

State of the News Media (2005) Project for Excellence in Journalism. Available online at http://www.stateofthenewsmedia.org/2005, accessed April 2006

State of the News Media (2006) Project for Excellence in Journalism. Available online at http://www.stateofthenewsmedia.org/2006, accessed April 2006

Sundar S.S. and Nass, C. (2001) Conceptualizing sources in online news, Journal of Communication, Vol. 51, No. 1 March

Sunstein C. (2001) Republic.com, Princeton, Princeton University Press

Walker, L. (2001) AOL Time Warner sites dominate data, Washington Post, 27 February.

Washington Internet Daily (2006) Newspapers want pay from Google, other news aggregators for excerpts, February 3. Available on LexisNexis

Welch, M. (1999) Is reliance on the AP draining the life from online news? Online Journalism Review, May 20. Available online at http://www.ojr.org/ojr/business/1017968544.php, accessed April 2006

White, C. (2003) Google news performing well, journalism.co.uk, 8 January. Available online at http://www.journalism.co.uk/news/story538.html, accessed April 2006

Wilbert, C. (2004) Also-ran Fox retools Web site, Atlanta Journal-Constitution, June 26. Available on Lexis-Nexis

Note on contributor
Chris Paterson is a senior lecturer and Director of the International Communication MA at the Institute of Communications Studies, University of Leeds, UK. He has published extensively on news agencies and the production and flow of news. He has edited, with A. Sreberny, International News in the 21st Century (John Libbey, 2004). Since receiving his PhD from the University of Texas, Paterson has taught at the Centre for Mass Communication Research at the University of Leicester, and at the Universities of Georgia and San Francisco in the US and Ulster and Leeds in the UK. Email: c.paterson@leeds.ac.uk

• This article is adapted from a research paper prepared for the Centre for International Communications Research at the University of Leeds.
-----------------------

[1] For a review of early critiques of the utopian perspective, see Aufderheide (1998).

[2] Pew Research Centre For The People and The Press, Internet News: More Log On, Tune Out (June 11, 2001) people-press.org

[3] This was based on a telephone survey of 3000 US adults

[4] White reports on data gathered by www.newsknife.com

[5] Use of the term ‘audience’ in the context of a nominally interactive medium is usually discouraged, but in the new media sector of international news, a broadcast model applies, both for producers and audiences, so news readers are audiences in a traditional sense. News is broadcast from the few to the many, and there no interactivity of consequence

[6] See Paterson 1998, 1996; Boyd-Barrett 1998. Analysis of news agencies remains sparse, and the wire service production process poorly understood. To date, there is only limited large-scale ethnographic research on news agencies, although its focus is only the television side of these institutions (Paterson 1996). There have been smaller ethnographic projects and important works on agencies employing other methodologies (Hjarvard 1995b; Read 1992; Boyd-Barrett and Thussu 1992; Fenby 1986; Boyd-Barrett 1980)

[7] Disney’s lack of interest in news also helped to consolidate the agency sector, when they sold Worldwide Television News the video agency with the longest history - to its new rival, Associated Press Television, in 1997 – leaving just AP and Reuters in the business of large scale multimedia international news delivery

[8] In television, broadcasters write their stories around the video these organizations offer, and if they are not offered compelling images, they will minimize or ignore an international story. Studies of television newsrooms have shown that the availability of visual images is an important factor in determining whether a foreign news story is carried (Cohen et al 1996; Helland, 1995; Rodriguez 1996; Molina 1990; Schlesinger 1987).

[9] Other sources, which might typically be used by journalists on such a story, were also retrieved and stored, including US State Department statements, the relevant output of the Iran News Agency, and the statements of a London-based Iranian dissident group. Although not a goal of the present study, it was evident that on stories such as this almost every original source consulted by the writer can be identified – stories can be fully deconstructed

[10] plagiarism.phys.virginia.edu. Many such programs are now available, and recent editions of Microsoft Word contain text comparison features. The software indicates strings of text in one document that are copied from another. The user can set the parameters of what constitutes ‘copying’. In this case, ‘copying’ was the duplication of a string of five or more words. After trial runs, this appeared the best compromise to avoid the counting of names or common expressions (i.e. ‘appeared to be’), but to catch short sentences, or significant portions of sentences, copied from wire services without alteration. The system is imperfect. Among the problems encountered were when lengthy names were used together in similar ways from story to story (i.e. ‘Judge Juan Guzman charged General Pinochet’), adding, erroneously, to the total of copied words.

[11] In each year of the study, the sample was determined over the course of 1-2 weeks by monitoring international news during periods when research assistants were available to gather stories, and identifying stories which:

1. were not about, and did not take place in, the US or UK, where the media being analysed are based

2. normally would have no strong relevance to either the US or the UK

3. were selected by the major news agencies and at least three major news outlets as being worthy of prominent display during the periods analysed

4. were about events occurring outside of major news hotspots; especially the Middle East.

5. were new and breaking stories of international significance

When qualifying stories were identified as many versions as could be gathered of Reuters, AP and AFP output related to the story were obtained – through various means – and each news service in the study was monitored. If they posted a story on the topic, it was electronically saved for analysis. Although some exceptions were necessary, we attempted to gather data for each story topic within the same period of about four hours (so iterations of a story published over many days are not part of this study)

[12] While both the New York Times and Washington Post have emerged as leading online news outlets in international rankings, the New York Times usually ranks ahead, and so was the only one of these included in this quantitative study. Figures on the Times’ international readership are not available, but one industry report claims that nearly ¾ (72 per cent) of the Times’ US online readership comes from outside of the New York area (Hitwise.com, June 1, 2005). While the portal Excite.com, owned by Ask.com parent IAC Media, had declined in popularity during the period of this study, it also remains important and was included in the quantitative portion of this research. Excite hyperlinks from its home page to lists of news from each of the following services – but frequently the stories of most, or all, will be the same wire copy as the first two lists, from the agencies themselves: AP, Reuters, New York Times, CBS, MSNBC, USA Today, Fox News

[13] Kutz and Herring (2005) found that 51 per cent of changes to stories in the three major news websites they observed added no new information, consisting just of ‘formatting, spelling, grammar corrections; stylistic polish; rhetorical/ideological rewording’

[14] The original 2001 survey consisted of 14 stories (distinct news story topics), but for the sake of this reanalysis and comparison to ten 2006 stories, the researcher cut those to the ten stories receiving the most coverage. This provides slightly different results from the original 2001 study.

[15] It is unclear if the remaining 32 per cent accounts for actual changes made by web portal editorial staff to agency stories, variations in identifying text or unrelated portions of the web page (which were not edited out when the original texts were gathered), or slight changes in the versions of the wire service story compared. It is likely a combination of those factors

[16] Although we sought to gather data on breaking stories within a short period to minimise the possibility of multiple versions of a news agency story confusing the results, this happened in some instances. In cases where a news portal’s story was clearly that of a news agency – because it was labelled as such and because the portal in question is known not to make editorial alterations, as with excite or altavista – but our analysis did not match the story fully with that of a news agency – it is assumed that the portal story is a later or earlier version of the news agency story which we did collect. 100 per cent news agency use was therefore recorded, despite our less than 100 per cent match

[17] Where a small portion of verbatim news agency material appears in the story of a news service which generally does original reporting, as with the New York Times, it is normally the result of that service using an extended quote from the wire service’s story. Increasingly, writers tend to acknowledge the news agency by name when using such quotes. It is not clear whether this has been instituted as a new requirement of news agency contracts with clients, or whether news services are simply becoming more cautious about attribution amid declining public trust in news and a string of high profile reporting scandals, especially at the New York Times.

[18] It is beyond the means of this paper to address the next logical question: who sets the agenda of the news agencies? This is a vital question, nonetheless, and has been addressed in some detail in this author’s earlier work and other studies of international news cited previously.